Onward! 2012 program

I am program chair of Onward! 2012. What do you want to see?

Here is one thing I am thinking about. I understand that in the real sciences, conferences often accept papers based solely on an abstract. That wouldn’t work for Onward, or even most of CS. Perhaps we should instead use the opposite of an abstract: a “concrete”. That is, an example. A common problem with big-idea papers is the lack of examples, making them quite difficult to understand. Perhaps we ought to require authors to submit an example up front in lieu of an abstract. That procedure might be beneficial both for the authors and the reviewers.

This entry was posted in General. Bookmark the permalink. Both comments and trackbacks are currently closed.


  1. Posted August 31, 2011 at 1:50 pm | Permalink

    One thing I’d like to see is a journal which had staff writers who worked with scientists (like biographers work with the people they are biographing) to write articles in a more understandable way. It would also be nice if journals helped scientists release any code they have written in an easy-to-use open-source way.

    • Posted September 1, 2011 at 1:50 am | Permalink

      In fact I’d say why not allow an experienced programmer to review the code and suggest changes, before releasing it?

  2. Chung-chieh Shan
    Posted September 1, 2011 at 8:52 pm | Permalink

    Many linguistics conferences ask for such “concretes”, and it works quite well, except we call them “abstracts”.

  3. Justin
    Posted September 2, 2011 at 5:55 am | Permalink

    A “paper” need not be limited by constraints of actual paper. Encourage experimentation with possibilities offered by media such as hypertext, or even more fancy things, to present ideas and data in a format more easily absorbed, and the option of not reading through introductions to a material already well known to parts of the readership.

  4. Sean McDirmid
    Posted September 2, 2011 at 7:12 pm | Permalink

    Innovation in a conference is difficult, you run the danger of appearing gimmicky. The paper/review format is hard to innovate on

    Perhaps take a look at what they do at CHI, almost all papers are accompanied by videos (UIST, SIGGRAPH also). Onward has a separate video track, perhaps it is more prudent today to combine the tracks (paper + highly recommended video). Video production is hard though, we often lack the expertise to do it decently. You could just force the issue now, it would be a small but meaningful step.

    Another idea is to involve submitters/PC/wider-community in new kinds of feedback loops more appropriate to today’s use of social media (reddit, stackoverflow, facebook). Perhaps, before formal PC review, have a wider community review rank submissions with like/+1 and dislike/-1, along with open comments. Perhaps these mean little to the PC (or maybe reviewers can only review them after the fact), but it would be interesting to see how community ranks correspond to PC ranks. Could we possibly crowdsource paper reviews in the future? Probably not, but it would be interesting to ponder.

    • Sean McDirmid
      Posted September 4, 2011 at 1:56 am | Permalink

      Replying to myself, there are two reasons why I don’t include a video in my current PL pubs. First, decent video production is expensive in time and resources. Second, the one time I have included video in a submission, I found that the PC didn’t bother watching it (I embedded the videos into my PDF, there was no secondary upload at that time for supporting materials).

      Strongly encouraging videos in a conference call means two things: first, my paper would be at a disadvantage if a video was appropriate; I wouldn’t submit to UIST without a video (though I haven’t gotten in yet :p). Second, if videos are encouraged, then it is implied that the PC will actually watch the videos for anything that makes it to the PC meeting. Then making a video doesn’t feel like wasted effort for the submission.

  5. Morgan
    Posted September 4, 2011 at 12:03 am | Permalink

    Multi-media ‘coordination’ languages like Max/MSP, PD, VVVV and nascent systems in Clojure (Overtone), Haskell, Pure

  6. Posted September 10, 2011 at 8:39 am | Permalink

    Wow, this is amazing news! I love the “example” idea, btw, but I have some sympathy with Sean’s comment too. But if you can’t innovate in Onward! where can you?

  • Subscribe to Blog via Email

    Enter your email address to subscribe to this blog and receive notifications of new posts by email.